💊 TimeCapsule-SLM

Complete AI-powered research & creative platform with DeepResearch

Generate Novel Ideas • Build AI Content • Enable Collaborative Knowledge Discovery

---

🔗 Quick Links

• 🚀 Key Features
• 🌐 Experience Live Now
• 🚦 How to Start
• 🐳 Docker Support- In Progress
• 🔗 How to Share
• 💬 Join Discord Community
• 🦙 Ollama Integration
• 🏠 LM Studio Integration
• 🔬 DeepResearch Usage
• 🎮 Playground Usage
• 📄 Project Structure
• 🚀 Powered By

---

🚀 **Key Features **

🧠 In-Browser RAG | 🔗 TimeCapsule Sharing | 📚 Knowledge Base | 🤖 Local LLM Support

🧠 In-Browser RAG Semantic search and Retrieval-Augmented Generation with your own documents—directly in your browser. No server, no data leaves your device. 📚 Knowledge Base Integration Upload PDFs, text, and images to build a private, searchable research knowledge base with intelligent document analysis.

🔗 TimeCapsule Sharing Export and load full research sessions as .timecapsule.json files. Instantly share, restore, or collaborate on research with anyone. 🤖 Local LLM Support Use Ollama, LM Studio, OpenAI, and Anthropic—privacy-first, cost-effective, and lightning fast.

---

🌐 ✨ Experience Live Now! ✨

🎯 Launch TimeCapsule-SLM →

🚀 Instant Access No downloads, no setup - just click and create! Professional-grade research and creative coding in your browser.

🤖 AI-Powered Ollama, LM Studio & API (OpenAI, Anthropic) integration for intelligent research analysis and creative code generation.

💊 What You Get:

🔬 DeepResearch TimeCapsule - Comprehensive AI-powered research platform
🎮 Playground - Execute TimeCapsules with creative coding
🧠 Triple AI Mode: Ollama, LM Studio and APIs (OpenAI, Anthropic)
⚙️ Custom Context Templates for personalized AI behavior
📱 Responsive Design that works on all devices
🔄 Seamless Navigation between research and creative modes
🔒 Privacy First with multiple local AI options

---

🚦 How to Start

🎯 Get Research-Ready in 5 Minutes

🌐 Option 1: Instant Online (Recommended) 🎯 Go to timecapsule.bubblspace.com 🔬 Click "DeepResearch" 🦙 Start Ollama (see setup below) 🤖 Pull a model: ollama pull qwen3:0.6b 📚 Add documents in Knowledge Manager 📝 Add research topics and click Generate

🐳 Option 2: Docker (Easy Deploy) 📁 Clone: git clone https://github.com/thefirehacker/TimeCapsule-SLM 📂 Navigate: cd TimeCapsule-SLM 🐳 Start: docker-compose --profile ai-enabled up -d 🌐 Access: http://localhost:3000 🦙 Pull model: docker exec timecapsule-ollama ollama pull qwen3:0.6b 🚀 Start researching!

💻 Option 3: Local Development 📁 Clone: git clone https://github.com/thefirehacker/TimeCapsule-SLM 📂 Navigate: cd TimeCapsule-SLM 🌐 Open: DeepResearch.html in browser 🦙 Setup Ollama (see integration guide) 🚀 Start researching!

🦙 Quick Ollama Setup (Essential for local AI)

# 1. Install Ollama from https://ollama.ai

# 2. Pull recommended model
ollama pull qwen3:0.6b

# 3. Start with CORS enabled (CRITICAL)
OLLAMA_ORIGINS="https://timecapsule.bubblspace.com/,http://localhost:3000" ollama serve

# 4. Connect in TimeCapsule-SLM

💡 Pro Tip: For best results, use Ollama with the qwen3:0.6b model. LM Studio and APIs (OpenAI, Anthropic) are also fully supported.

---

🔗 How to Share

🤝 Collaborate & Share Research Instantly

📤 Export TimeCapsule 🔬 Complete your research in DeepResearch 💾 Click "Export TimeCapsule" button 📁 Save .timecapsule.json file 🤝 Share with colleagues or save for later Perfect for: Research collaboration, session backup, knowledge sharing

📥 Load TimeCapsule 📂 Click "Load TimeCapsule" button 🗂️ Select .timecapsule.json file ⚡ Instantly restore topics and research output 🔄 Continue where you left off Perfect for: Resuming sessions, importing shared research, team collaboration

🎯 TimeCapsule Features

• 🔄 Complete Session Restore - All topics, research results, and notes
• 📊 Multi-Tab Support - Research, Sources, and Notes tabs preserved
• 🤝 Team Collaboration - Share research across teams instantly
• 💾 Session Backup - Never lose your research progress
• 🌐 Cross-Platform - Works on any device with TimeCapsule-SLM

---

🦙 Ollama Integration

🎯 Local AI Power + Privacy First

Complete platform-specific setup guides for macOS, Linux & Windows

🚀 Why Ollama? 🔒 Fully Private - All processing happens locally 💰 Zero API Costs - No usage fees or limits ⚡ Lightning Fast - Optimized GGUF models 🎛️ Model Library - Easy model management 🌐 REST API - Simple integration

🛠️ Setup Requirements Ollama App - Download from ollama.ai AI Model - Any compatible GGUF model CORS Enabled - CRITICAL for web access Port 11434 - Default Ollama server port

---

🍎 macOS Setup Guide

📥 Step 1: Install Ollama

# Method 1: Direct download (recommended)
# Download from https://ollama.ai and install .app

# Method 2: Homebrew
brew install ollama

🤖 Step 2: Pull a Model

# Recommended: Fast and efficient
ollama pull qwen3:0.6b

🔧 Step 3: Start with CORS (CRITICAL)

# Kill any existing processes first
pkill -f ollama

# Start with CORS enabled (for testing)
OLLAMA_ORIGINS="*" ollama serve

# For production (recommended)
OLLAMA_ORIGINS="https://timecapsule.bubblspace.com/,http://localhost:3000" ollama serve

🔧 macOS Troubleshooting

❌ "Operation not permitted" Error:

# Method 1: Use sudo
sudo pkill -f ollama

# Method 2: Activity Monitor (GUI)
# 1. Open Activity Monitor (Applications → Utilities)
# 2. Search for "ollama"
# 3. Select process and click "Force Quit"

# Method 3: Homebrew service (if installed via brew)
brew services stop ollama
brew services start ollama

❌ CORS Issues:

# 1. Stop Ollama completely
sudo pkill -f ollama

# 2. Wait 3 seconds
sleep 3

# 3. Start with CORS
OLLAMA_ORIGINS="*" ollama serve

# 4. Test connection
curl http://localhost:11434/api/tags

---

🐧 Linux Setup Guide

📥 Step 1: Install Ollama

# Official installer (recommended)
curl -fsSL https://ollama.ai/install.sh | sh

# Or download directly from https://ollama.ai

🤖 Step 2: Pull a Model

# Recommended model
ollama pull qwen3:0.6b

🔧 Step 3: Configure CORS with systemctl (CRITICAL)

For systemd-based Linux distributions (Ubuntu, Debian, CentOS, etc.):

# 1. Stop any running Ollama instances
ps aux | grep ollama
sudo pkill -f ollama

# 2. Edit the ollama service configuration
sudo systemctl edit ollama.service

# 3. Add the following environment variables:
[Service]
Environment="OLLAMA_HOST=0.0.0.0"
Environment="OLLAMA_ORIGINS=*"

# For production, use specific origins:
# Environment="OLLAMA_ORIGINS=https://timecapsule.bubblspace.com/,http://localhost:3000"

# 4. Save and exit the editor (Ctrl+X, then Y, then Enter)

# 5. Reload systemd and restart ollama service
sudo systemctl daemon-reload
sudo systemctl restart ollama.service

# 6. Enable auto-start on boot (optional)
sudo systemctl enable ollama.service

# 7. Verify the service is running
sudo systemctl status ollama.service

# 8. Test the connection
curl http://localhost:11434/api/tags

Alternative: Manual start (if not using systemd):

# Stop any existing processes
sudo pkill -f ollama

# Start manually with CORS
OLLAMA_ORIGINS="*" ollama serve

# Or for production:
# OLLAMA_ORIGINS="https://timecapsule.bubblspace.com/,http://localhost:3000" ollama serve

🔧 Linux Troubleshooting

❌ Service Issues:

# Check service logs
sudo journalctl -u ollama.service -f

# Restart service
sudo systemctl restart ollama.service

# Check service status
sudo systemctl status ollama.service

❌ Permission Issues:

# Stop with elevated permissions
sudo pkill -f ollama

# Check for lingering processes
ps aux | grep ollama

# Force kill if needed
sudo kill -9 $(pgrep ollama)

❌ CORS Configuration:

# Verify environment variables are set
sudo systemctl show ollama.service | grep Environment

# If not set, re-edit the service:
sudo systemctl edit ollama.service
# Add Environment variables as shown above
sudo systemctl daemon-reload
sudo systemctl restart ollama.service

📚 Reference: Ollama CORS Configuration Guide

---

🪟 Windows Setup Guide

📥 Step 1: Install Ollama

# Download from https://ollama.ai and install the .exe
# Or use package manager (if available)

🤖 Step 2: Pull a Model

# Open Command Prompt or PowerShell
ollama pull qwen3:0.6b

🔧 Step 3: Start with CORS (CRITICAL)

# Method 1: Stop existing processes
taskkill /f /im ollama.exe

# Method 2: Start with CORS (Command Prompt)
set OLLAMA_ORIGINS=* && ollama serve

# Method 3: Start with CORS (PowerShell)
$env:OLLAMA_ORIGINS="*"; ollama serve

# For production (specific origins):
# $env:OLLAMA_ORIGINS="https://timecapsule.bubblspace.com/,http://localhost:3000"; ollama serve

🔧 Windows Troubleshooting

❌ Process Issues:

# Method 1: Task Manager (GUI)
# 1. Open Task Manager (Ctrl+Shift+Esc)
# 2. Look for "ollama.exe" in Processes tab
# 3. Right-click and select "End task"

# Method 2: Command line
taskkill /f /im ollama.exe

# Method 3: Find by port
netstat -ano | findstr :11434
# Note the PID and kill it:
taskkill /f /pid <PID>

❌ CORS Issues:

# 1. Stop all ollama processes
taskkill /f /im ollama.exe

# 2. Wait 3 seconds
timeout /t 3

# 3. Start with CORS
$env:OLLAMA_ORIGINS="*"; ollama serve

# 4. Test connection (if curl is available)
curl http://localhost:11434/api/tags

❌ Environment Variables:

# Set permanently (requires restart)
setx OLLAMA_ORIGINS "*"

# Set for current session only
$env:OLLAMA_ORIGINS="*"

---

🎯 Universal Commands & Verification

🧪 Test Your Setup

# 1. Check if Ollama is running
curl http://localhost:11434/api/tags

# 2. List installed models
ollama list

# 3. Test model response
curl http://localhost:11434/api/generate -d '{
  "model": "qwen3:0.6b",
  "prompt": "Hello",
  "stream": false
}'

📦 Recommended Models

Model	Size	Best For	Performance
qwen3:0.6b	~400MB	Fast responses, testing	🌟🌟🌟🌟⭐
qwen2.5:3b	~2GB	Balanced quality/speed	🌟🌟🌟🌟🌟
llama3.2:3b	~2GB	General purpose	🌟🌟🌟⭐⭐

# Pull additional models:
ollama pull qwen2.5:3b
ollama pull llama3.2:3b

🆘 Universal Reset (All Platforms)

If everything fails, complete reset:

# 1. Stop all Ollama processes
# macOS/Linux: sudo pkill -f ollama
# Windows: taskkill /f /im ollama.exe

# 2. Wait 5 seconds
sleep 5  # macOS/Linux
# timeout /t 5  # Windows

# 3. Start fresh with CORS
OLLAMA_ORIGINS="*" ollama serve
# Windows PowerShell: $env:OLLAMA_ORIGINS="*"; ollama serve

# 4. Pull a model (in new terminal)
ollama pull qwen3:0.6b

# 5. Test setup
curl http://localhost:11434/api/tags

💡 Pro Tips:

• Linux Users: Use systemctl for persistent CORS configuration
• macOS Users: Use Activity Monitor for stubborn processes
• Windows Users: Use Task Manager or PowerShell for process management
• All Platforms: Use OLLAMA_ORIGINS="*" for testing, then restrict to specific domains
• Always verify your setup with: curl http://localhost:11434/api/tags

🌐 Custom Ollama URL (Local Builds Only)

⚠️ Note: Custom URLs only work in local builds, not hosted version.

Easy Setup with ollama-custom.js:

1. Edit Configuration File: Open ollama-custom.js in the root directory
2. Add Your IPs: Replace the example IPs with your actual Ollama servers

   customIPs: [
     "http://10.0.1.69:11434",      // Your first Ollama server
     "http://192.168.1.200:11434",  // Your second Ollama server  
     "http://172.16.0.50:9434"      // Your third Ollama server
   ]

3. Save and Refresh: Save the file and hard refresh your browser (Ctrl+Shift+R)

Use in App: Click "Connect Ollama" → accept agreement → Enter custom URL in popup

• DeepResearch: Click "🦙 Connect Ollama"
• Playground: Click "Connect AI" → Select Ollama

Examples: http://192.168.1.100:11434, http://localhost:9434, https://ollama.mydomain.com

---

🏠 LM Studio Integration

🎯 Local AI Power + No API Costs

🚀 Why LM Studio? 🔒 Fully Private - All processing happens locally 💰 Zero API Costs - No usage fees or limits ⚡ Fast Response - Direct local connection 🎛️ Model Control - Use any compatible model 🌐 OpenAI Compatible - Standard API format

🛠️ Setup Requirements LM Studio App - Download from lmstudio.ai Compatible Model - Any chat-capable model Local Server - LM Studio server on port 1234 CORS Enabled - Allow cross-origin requests

📋 Quick Setup Guide

🚨 KEY REQUIREMENT: You MUST enable CORS in LM Studio for TimeCapsule-SLM to connect.

Step 1: 📥 Download LM Studio from lmstudio.ai and install it
Step 2: 🤖 Download a Model - Search for models like Qwen3 0.6B Step 3: 🚀 Start Local Server - Click "Start Server" in LM Studio (port 1234)
Step 4: ⚙️ Enable CORS - IMPORTANT: In LM Studio → Settings → Server → Enable "CORS"
Step 5: 🔄 Restart Server - Stop and restart the LM Studio server
Step 6: 💊 Connect in TimeCapsule - Select "🏠 LM Studio" from AI provider dropdown
Step 7: 🔌 Click Connect - TimeCapsule will auto-detect your model

🎯 Recommended Models

Model	Size	Best For	Performance
Qwen3 0.6B	~500MB	Research analysis, detailed coding responses	🌟🌟🌟🌟🌟

---

🔬 DeepResearch TimeCapsule Usage

🧠 AI-Powered Research Generation

📊 Research Workflow

1. 📝 Add Topics - Define research areas with descriptions
2. 🎯 Select Type - Choose from Academic, Market, Technology, Competitive, Trend, Literature
3. 📏 Set Depth - Pick Overview, Detailed, or Comprehensive analysis
4. 🤖 Generate Research - AI creates structured, professional reports and TimeCapsules
5. 📤 Export Results - Download as .timecapsule.json files for sharing

🎯 Research Types Explained

• 📚 Academic - Scholarly analysis with citations and methodology
• 📈 Market - Industry trends, competition, and market analysis
• 🔧 Technology - Technical deep-dives and implementation insights
• 🏢 Competitive - Competitor analysis and market positioning
• 📊 Trend - Emerging patterns and future predictions
• 📖 Literature - Comprehensive literature reviews and surveys

---

🎮 Playground Usage

🎨 Creative Coding with AI Assistance

---

📄 Project Structure

📁 Clean, Organized Architecture

💊 Core Application Files File Description DeepResearch.html DeepResearch TimeCapsule studio Playground.html Playground creative AI environment Canvas.html Creative CodingEnvironment index.html Main platform homepage Script01.js Utility functions and helpers

📋 Documentation & Assets File Description README.md This comprehensive guide CREDITS Algorithm attributions LICENSE Apache 2.0 License lib/ Assets and design templates

📂 Library Structure

• lib/agent/ - Canvas AI agents
• lib/AIAssistant/ - AI backend integration
• lib/Designs/ - Creative coding templates
• lib/Pages/ - Component libraries
• lib/Media/ - Images and assets

---

🚀 Powered By

Proudly Supported by BubblSpace

Building the future of AI-powered creativity and collaboration

---

🧙‍♂️ Created with ❤️ by FireHacker

🌍 Made for researchers, creators, developers, and digital artists worldwide

🐦 Follow @thefirehacker

⭐ STAR THIS PROJECT ⭐ Help us reach 100 stars!

🎮 Discord Community

---

💬 Support & Community

🎧 Discord Community Real-time help & discussions Connect with fellow researchers! 💬 discord.gg/ExQ8fCv9

📧 Email Support Direct technical assistance Professional support team 📧 support@bubblspace.com

🐛 Report Issues Bug reports & feature requests Help improve the platform 🔧 GitHub Issues

📚 Documentation Complete guides & tutorials Everything you need to know 📖 View Docs • 🐳 Docker

🆘 Get Help With:

🔧 Setup & Installation • 🤖 AI Integration • 🔬 Research Workflows • 📚 Document Management • 🎮 Creative Coding • 🔄 TimeCapsule Sharing • 🐛 Troubleshooting

---

⭐ LOVE THIS PROJECT? GIVE IT A STAR! ⭐

🎯 Help us reach 100 stars and unlock new features!

⚡ Just one click makes a huge difference!

🙏 Your star helps more developers discover TimeCapsule-SLM and supports continued development!

🤝 Join Our Growing Community

🌟 Star Gazers Join our amazing community of developers

🍴 Contributors Be part of the development journey

📈 Project Stats Growing stronger every day

---

💫 Thank you for being part of the TimeCapsule-SLM community! Together, we're revolutionizing AI-powered research and creativity. 💫